trainer_state.json

{
  "best_metric": 1.0984911918640137,
  "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_mnli_256/checkpoint-15340",
  "epoch": 10.0,
  "global_step": 30680,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.9e-05,
      "loss": 1.0987,
      "step": 3068
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.098777413368225,
      "eval_runtime": 17.5509,
      "eval_samples_per_second": 559.231,
      "eval_steps_per_second": 4.387,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-05,
      "loss": 1.0987,
      "step": 6136
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.0986140966415405,
      "eval_runtime": 17.4915,
      "eval_samples_per_second": 561.131,
      "eval_steps_per_second": 4.402,
      "step": 6136
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.7e-05,
      "loss": 1.0986,
      "step": 9204
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.0988407135009766,
      "eval_runtime": 17.529,
      "eval_samples_per_second": 559.93,
      "eval_steps_per_second": 4.393,
      "step": 9204
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 1.0986,
      "step": 12272
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.098751425743103,
      "eval_runtime": 17.6284,
      "eval_samples_per_second": 556.771,
      "eval_steps_per_second": 4.368,
      "step": 12272
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.5e-05,
      "loss": 1.0986,
      "step": 15340
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0984911918640137,
      "eval_runtime": 17.4397,
      "eval_samples_per_second": 562.795,
      "eval_steps_per_second": 4.415,
      "step": 15340
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 1.0986,
      "step": 18408
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.0986493825912476,
      "eval_runtime": 17.4712,
      "eval_samples_per_second": 561.781,
      "eval_steps_per_second": 4.407,
      "step": 18408
    },
    {
      "epoch": 7.0,
      "learning_rate": 4.3e-05,
      "loss": 1.0986,
      "step": 21476
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3273560876209883,
      "eval_loss": 1.0989376306533813,
      "eval_runtime": 17.4516,
      "eval_samples_per_second": 562.412,
      "eval_steps_per_second": 4.412,
      "step": 21476
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.2e-05,
      "loss": 1.0986,
      "step": 24544
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0985747575759888,
      "eval_runtime": 17.5984,
      "eval_samples_per_second": 557.721,
      "eval_steps_per_second": 4.375,
      "step": 24544
    },
    {
      "epoch": 9.0,
      "learning_rate": 4.1e-05,
      "loss": 1.0986,
      "step": 27612
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.3544574630667346,
      "eval_loss": 1.0985636711120605,
      "eval_runtime": 17.4957,
      "eval_samples_per_second": 560.994,
      "eval_steps_per_second": 4.401,
      "step": 27612
    },
    {
      "epoch": 10.0,
      "learning_rate": 4e-05,
      "loss": 1.0986,
      "step": 30680
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.31818644931227713,
      "eval_loss": 1.0986591577529907,
      "eval_runtime": 17.4485,
      "eval_samples_per_second": 562.514,
      "eval_steps_per_second": 4.413,
      "step": 30680
    },
    {
      "epoch": 10.0,
      "step": 30680,
      "total_flos": 1.0122133036335104e+17,
      "train_loss": 1.0986432768064496,
      "train_runtime": 20712.8857,
      "train_samples_per_second": 947.965,
      "train_steps_per_second": 7.406
    }
  ],
  "max_steps": 153400,
  "num_train_epochs": 50,
  "total_flos": 1.0122133036335104e+17,
  "trial_name": null,
  "trial_params": null
}